You are on page 1of 16

5.

6 Poisson Distribution and the Poisson Process


Some experiments result in counting the numbers of
particular events occur in given times or on given
physical objects, such as :
a line segment, an area, a volume, or perhaps a
piece of a metal.
For example, we may want to count
the number of flaws in 100 feet of a wire,
number of white cells in a drop of blood.
number of misprint on a page of a book.
Each count can be looked upon as a random variable
associated with an approximate Poisson process. The
experiments are called Poisson experiments.

1
Properties of Poisson
Process
The number of outcomes occurring in one time interval
or specified region is independent of the number that
occurs in any other disjoint time interval or region of
space.
The probability that a single outcome will occur during
a very short time interval or in a small region is
proportional to the length of the time interval or the
size of the region and does not depend on the number
of outcomes occurring outside this time interval or
region. P( X (t , t t ) 1) t (t )

2
The probability that more than one outcome will
occur in such a short time interval or fall in such
a small region is negligible.

P( X (t, t t ) j) (t )
j 2
Let X be the number of outcomes occurring during
a Poisson experiment, then X is called a Poisson
random variable.
The probability distribution of X is called
the Poisson distribution. The possible values
for X are 0, 1, 2, 3, .

3
Poisson Distribution
Let X be a Poisson random variable, representing the
number of outcomes occurring in a given time
interval or specified region denoted by t. Then
e t ( t ) x
p(x; t)= P(X = x) = , x = 0, 1, 2, .,
x!
Where is the average number of outcomes per unit
time or per unit region. (rate of the occurrence of the
outcomes)
Table A.2 contains Poisson probability sum
r
e t (t ) k
P(X r) = p ( x; t ) =
r

x 0 k 0 k!

4
Example 5.19, page 137 During a laboratory
experiment the average number of radioactive
particles passing through a counter in 1 millisecond is
4. What is the probability that 6 particles enter the
counter in a given millisecond.
Solution: = 4, t = 1, t = 4, x = 6.

p(6;4)= e 4 (4) 6
6!
=P(X 6) P(X 5)
= 0.1042
Example

5
Example 5.20, page 137 Ten is the average number
of oil tankers arriving each day at a certain port city.
The facilities at the port can handle at most 15 tankers
per day. What is the probability that on a given day
tankers have to be turned away?
Solution: Let X be the number of tankers arriving
each day.
= 10, t = 1, t = 10. Find P(X > 15).
P(X > 15) = 1 P(X 15) = 1 P(15; 10)
15
= 1 p ( x;10)
x 0

= 0.0487.

6
Theorem 5.5 The mean and variance of the Poisson
distribution p(x; t) both have value t.
=t , 2
=t
Proof: (See Appendix A26)

et (t ) k
(t ) k 1
k P( X k ) k (t )et t
k 0 k 1 k! k 1 (k 1)!

e t ( t ) k
E( X ) k
2 2 2
2 2

k 1 k!

(t ) k 1
( t )e t

k 1
k
(k 1)!
2


(t ) k 1
(t ) k 1
( t )e t

k 1
( k 1)
(k 1)!
( t ) e t

k 1 ( k 1)!
2


(t ) k 2
( t ) e
2 t

k 2 ( k 2)!
t 2 t

7
Example
The white blood-cell count of a healthy individual
can average as low as 6000 per cubic millimeter of
blood. To detect a white-cell deficiency, a 0.001
cubic millimeter of blood is taken and the number of
white cells X is found.
(a) How many white cells are expected in a healthy
individual?
= 6000, t = 0.001, t = 6.
=t =6
In a healthy individual, we would expect, on
average, to see six white cells per 0.001 cubic
millimeter drop of blood.
8
(b) What is the probability of finding at most one
white cell?
P(X 1) = 0.0174
The chance of observing none or one white cell per
0.001 cubic millimeter drop of blood of a healthy
individual is only 0.0174.
In other words, in 1000 healthy individuals, only
about 17 of them would provide the blood drops that
contain none or one white cell.
Hence, if either none or one white cell found in a
blood drop of an individual, we would think that there
is good evidence that this individual is not healthy, or
there is a white-cell deficiency in his or her blood,
although there is chance for error.
9
Theorem 5.6 Let X be a binomial random variable with
probability distribution b(x; n, p). If when n , p 0 and
= np remains constant, then
b(x; n, p) p(x; ) as n
That is to say: if lim npn (>0) ,then
n
k
lim Cnk pn (1 pn ) n k e k 0,1,2,
k
n k!
Proof:
(npn / n) k np
Cnk pn (1 pn ) n k n(n 1)...( n k 1) (1 n ) n k
k

k! n
1 k 1 (npn ) k np
n
npn
nk
1 (1 )...(1 ) [(1 n ) npn ] n
n n k! n
n k
e
k!
10
Application: if X ~B(n, p)when n >10p < 0.1
we have
(np) k np
P( X k ) e
k!

Example: The newborns chromosome( )


abnormality rate is 1% according to the experience. Let X
be the number of chromosome abnormality among 100
newborns. Please use the principles of Binominal and
Poisson distribution to calculate the probability
distribution of random variable X.
n=100, X=0,1,2,,100, p=0.01, = np=1.

11
P(X)
X
Binominal Poisson
0 0.3660 0.3679
1 0.3697 0.3679
2 0.1849 0.1839
3 0.0610 0.0613
4 0.0149 0.0153
5 0.0029 0.0031
6 0.0005 0.0005
7 0.0001 0.0001
8 0.0000 0.0000
1.0000 1.0000

12
Example
Example 2
A computer hardware company manufactures a particular type
of microchip. There is a 0.1% chance that any given microchip
of this type is defective, independent of the other microchips
produced. Determine the probability that there are at least 2
defective microchips in a shipment of 1000.
Solution: Let X be the number of defectives in a shipment of
1000. Then X has the distribution b(x; 1000, 0.001).

P( X 2) 1 P( X 0) P( X 1)
1 C1000
0
(0.999)1000 C1000
1
(0.001)(0.999)999
0.264211
13
Solution: since n is large, p is small, an
approximate model for X is the Poisson model
with np 1

P( X 2) 1 P( X 0) P( X 1)
e
1 e
1
1 2e 1 0.2642411
14
Example 3

The number of visits per minute to a particular Website


providing news and information can be modeled using a Poisson
distribution with mean 5. The Website can only handle 20 visits
per minute and will crash if this number of visits is exceeded.
Determine the probability that the site crashes in the next minute.
Solution: Let X be the number of visits to the site in the next
minute. Then X has the distribution p(x; 5).

P( X 20) 1 P( X 20)
20
5 x e 5
1
x 0
x!
0 15
Example 4

Let X be the number of some animals eggs which are laid


r
randomly, and the probability is given by p( X r ) er .
r!
Suppose that the probability of every eggs hatching a little
animal is p and each egg can hatch out a little animal
independently. If Y stands for the number of the animals
offspring, what is the cumulative distribution of Y?

16

You might also like